Laser imaging systems are extensively used to generate images in applications such as xerographic printing, mask and maskless lithographic patterning, laser texturing of surfaces, and laser cutting machines. Laser printers often use a raster optical scanner (ROS) that sweeps a laser perpendicular to a process direction by utilizing a polygon or galvo scanner, whereas for cutting applications laser imaging systems use flatbed x-y vector scanning.
One of the limitations of the laser ROS approach is that there are design tradeoffs between image resolution and the lateral extent of the scan line. These tradeoffs arising from optical performance limitations at the extremes of the scan line such as image field curvature. In practice, it is extremely difficult to achieve 1200 dpi resolution across a 20″ imaging swath with single galvanometers or polygon scanners. Furthermore, a single laser head motorized x-y flatbed architecture, ideal for large area coverage, is too slow for most high speed printing processes.
For this reason, monolithic light emitting diode (LED) arrays of up to 20″ in width have an imaging advantage for large width xerography. Unfortunately, present LED arrays are only capable of offering 10 milliWatt power levels per pixel and are therefore only useful for some non-thermal imaging applications such as xerography. In addition, LED bars have differential aging and performance spread. If a single LED fails it requires the entire LED bar be replaced. Many other imaging or marking applications require much higher power. For example, laser texturing, or cutting applications can require power levels in the 10 W-100 W range. Thus LED bars cannot be used for these high power applications. Also, it is difficult to extend LEDs to higher speeds or resolutions above 1200 dpi without using two or more rows of staggered heads.
Higher power semiconductor laser arrays in the range of 100 mW-100 Watts do exist. Most often they exist in a 1D array format such as on a laser diode bar often about 1 cm in total width. Another type of high power directed light source are 2D surface emitting VCSEL arrays. However, neither of these high power laser technologies allow for the laser pitch between nearest neighbors to be compatible with 600 dpi or higher imaging resolution. In addition, neither of these technologies allow for the individual high speed control of each laser. Thus high power applications such as high power overhead projection imaging systems, often use a high power source such as a laser in combination with a spatial light modulator such as a DLP™ chip from Texas Instruments or liquid crystal arrays.
The DLP-based approach, although appealing in many respects, has several disadvantages. First, the high power laser has to be fully on, even for low color saturation images that require only a small fraction of the laser energy to form. Most of the laser energy is diverted into a beam dump and wasted, leading to poor energy efficiency and reduced laser lifetime. Second, micromirrors absorb part of the laser energy, and can only handle a limited amount of optical power before they deform and fail. This limitation limits the print speed and long term reliability of the imaging system. Third, micro-mirrors are mechanical components that need to be carefully aligned to the rest of the optical system. Such mechanical switching components add complexity and cost.
What is needed is laser based imaging approach with high total optical power well above the level of 1 Watt that is scalable across large process widths in excess of 20″ as well as having achievable resolution greater than 1200 dpi and allows high resolution high speed imaging in a single pass.